Summary: Bipolar Disorder (BP) is a psychiatric illness with high morbidity, affecting from 1 to 4 % of all human populations. Most patients experience continued difficulty with psychiatric symptoms, lost productivity and functionality. Despite extensive genetic studies of the bipolar phenotype, and some interesting linkage and association findings identification of specific genes underlying the disorder have been difficult. The search for the genetic loci involved in BP has likely been hampered by the complexity of the illness and difficulty in defining the BP spectrum. One of the main problems in elucidating the genetic architecture of BP is the fact that the BP categorical diagnosis appears to be a poor predictor of correlation between the phenotype and the specific genotypic variants which contribute to an individual's risk of developing BP. Hence, indicators of processes mediating between genotype and phenotype (endophenotypes) are necessary to resolve the conflicting diagnostic status of family members in genetic studies of BP. Such biomarkers represent a window into the genetically influenced biological process underlying BP. Endophenotypes may represent a simpler and tractable etiology that can be quantitatively measured and analyzed with powerful analysis strategies not readily available for qualitative phenotypic markers such as diagnostic category. We have conducted a pilot study to investigate a quantitative measure of anxiety as a candidate endophenotype for BPI in BP families, by analyzing the heritability, genetic correlation and association with BPI. We found that quantitative measurements of state and trait anxiety are highly heritable, share some genetic factors and that the anxiety trait is associated with BPI. The current proposal intends to further examine anxiety trait in a larger sample of BPI extended pedigrees from a similar genetic and cultural environment (Costa Rican Central Valley population). We will ascertain 30 new extended pedigrees (600 individuals) with at least one individual diagnosed with BPI (by consensus diagnosis based on DSMIV) and 60 healthy unrelated controls. Quantitative anxiety will be evaluated by using the State-Trait Anxiety Inventory and additional self-rated anxiety related personality scales will be used to characterize anxiety as a potential. We will test whether anxiety fulfill the validation criteria to be an endophenotype for BPI by using specialized statistical analyses provided by SOLAR Software Package. We will collect blood specimens and beyond the scope of the current proposal: a complete genome screen will be conducted for all subjects (660 new individuals and 566 old subjects from our pilot study,) and QTL linkage and association analyses as well. We hypothesize that we will identify regions containing genes specific for these quantitative measures in the sample, as well as specific candidate genes of major effect underlying the anxiety component of BPI.